Since the introduction of lightweight portable computers (laptops, notebooks), a great deal of attention has been focused on the development of wireless computer networks (Wireless Local Area Network, WLAN). Thanks to standardization in the field of LANs, it is comparatively easy to find systems that will still be upgradable even in a few years' time. Around 70% of all computers connected to networks are compliant with the Institute of Electrical and Electronics Engineers (IEEE) 802.3 (Ethernet) and IEEE 802.5 (Token Ring) standards. Connection is normally over a permanent wireline link. The problems that can occur are the surfacing of mechanical defects (corrosion) after a few years and violations of rules on radiated interference. It is difficult to adapt these networks to cope with changing office conditions. Mobile network nodes are not possible.
The obvious approach is to leave out the cable entirely. This idea is almost as old as the concept of the so-called ALOHA system, which used radio to connect terminals to their processing computers. The newer WLANs work with the most up-to-date radio technology. Data is encrypted and extensive error-protection mechanisms are available. Integrity of data is also guaranteed. Just like wireline LANs, WLANs can be divided into different architectures and performance categories. Many companies offer products for wireless point-to-point connections, but only very few build LANs for multipoint communication. Today wireless networks use spread-spectrum, narrowband microwave or infrared signals for transmission. Because of legal regulations, networks using spread-spectrum and narrowband microwave cannot be operated in most countries unless special authorization has been given. The only exemption of this is operation in a license exempt band, e.g., Industrial, Scientific and Medical (ISM) band, where under a set of given rules for channelization and emitted power operation of radio equipment is allowed, generally. An example of this is the 2.4 GHz band, where the first WLANs have been positioned.
A WLAN need not to be organized centrally, and instead may have a completely distributed architecture with a dynamic allocation of network and network node identifiers. In contrast to a wireline network, WLANs using the same radio channel cannot be separated from one another. Overlapping can occur. Another problem with radio channels is the range restriction. Mobile WLAN nodes and unfavourable propagation characteristics can cause the fragmentation of a network.
As a result of the channel characteristics and the applications for which WLANs are advantageous, WLANs have to be frequently operated in an overlapped fashion. If the radio range of some of the stations in network A should overlap with some of the stations in network B then these members share the transmission medium and its transmission capacity in the area where the overlapping occurs. An overlapping of networks produces two effects: First, the senders in the different WLANs use the same frequency band, thereby increasing the occurrences of interference. As a result, optimal use of the frequency band is no longer possible because not all the stations are able to receive from each other (hidden stations) and therefore can cause interference to each other. Second, a station receives data packets from several WLANs with different WLAN identifiers (LIDs). All received data packets are evaluated, and only those with its own LIDs are accepted. As a result, there is a decrease in the maximum possible data transmission capacity and consequently also the data transmission rate in this area.
In view of this problem, it is thus an object of the invention to improve the performance of wireless communication systems by enhancing the coexistence with other wireless communication systems.
As an example for a WLAN standard, IEEE 802.11 has been designed for use in Industrial, Scientific and Medical (ISM) bands. The Federal Communications Commission (FCC) in the USA prescribed maximum power levels only, band edge interference, and the requirement to use spread spectrum in order to minimize interference with already existing communication systems. It designated the frequency bands 902-928 MHz, 2400-2483.5 MHz and 5725-5850 MHz to these ISM bands.
IEEE 802.11 has defined two Physical Layer (PHY) standards for the 2.4 GHz ISM band: one using the Frequency Hopping Spread Spectrum (FHSS) technique, and one using the Direct Sequence Spread Spectrum (DSSS) technique. An alternative is the specification of an Infrared (IR) physical layer. IEEE 802.11 stations using any-of the three technologies operate at a data rate of 1 Mbit/s (optionally 2 Mbit/s) and at 11 Mbit/s according to standard IEEE 802.11 b. Recently work has been finished on the specification of a 20 Mbit/s PHY at 5 GHz. Target frequencies are those opened by the FCC in 1997 for Unlicensed Information Infrastructure Networks (U-NII).
The protocols of IEEE 802.11 are specified for slowly moving stations, usually indoors but not limited to this, communicating among each other (Ad Hoc Mode) or with stations beyond their direct communication range with the support of an infrastructure (Infrastructure Mode). The communication is packet-oriented.
The IEEE 802.11 MAC protocol provides two types of service: asynchronous and contention-free. The asynchronous type of service is provided by the Distributed Coordination Function (DCF), which implements as the basic access method the Carrier Sense Multiple Access (CSMA) with Collision Avoidance (CA) protocol. The contention-free type of service is provided by the Point Coordination Function (PCF), which basically implements a polling access method. Unlike the DCF, the implementation of the PCF is not mandatory. Furthermore, the PCF itself relies on the asynchronous service provided by the DCF.
The time between two frames is called Interframe Space (IFS). In order to determine whether the medium is free, a station has to use the carrier sense function for a specified IFS. The standard specifies four different IFSs, which represent three different priority levels for the channel-access. The shorter the IFS, the higher the priority. The IFSs are specified as time gaps on the medium and are independent of the channel data rate. Owing to the different characteristics of the different PHY specifications, the IFS time durations are specific for each PHY.
According to the DCF, a station must sense the medium before initiating the transmission of a packet. This mechanism is schematically depicted in FIG. 1.
FIG. 1 schematically depicts the transmissions of frames 101a. . . 101e of five stations 102a . . . 102e, where time proceeds in each row of FIG. 1 from left to right. Because the frames 101a . . . 101e have to be transmitted on the same shared transmission medium, a CSMA/CA protocol is obeyed by each of said stations 102a . . . 102e. The first row of FIG. 1 shows the transmission of a frame 101a by station 102a. As indicated by the vertical arrows 103b . . . 103d in the second, third and fourth row of FIG. 1, data packets arrive at stations 102b, 102c and 102d, so that these stations require to access said shared transmission medium to transmit the arriving data packets. Said three stations 102b, 102c and 102d now start sensing the medium. If the transmission medium is sensed as being busy, the transmission of stations 102b, 102c and 102d is deferred and a backoff process is started for each station, wherein said backoff process is only started after a DCF IFS (DIFS) period 104-1 during which the medium is determined to be idle for the duration of the DIFS. Specifically, each station computes a random number uniformly distributed between zero and a maximum called Contention Window (CW). The random number is multiplied by the slot time, resulting in the backoff interval used to set the backoff timer. In FIG. 1, said backoff intervals 105b . . . 105e are schematically depicted, wherein already elapsed backoff time slots are depicted as white boxes and remaining backoff time slots are depicted as gray boxes. For instance, in the third row of FIG. 1, the backoff interval of station 102c only consists of four time slots, and after the duration of said time slots, station 102c starts the transmission of a frame 101c. 
The backoff timers 105b . . . 105e are decremented only when the medium is idle, whereas they are frozen when another station is transmitting. This can be spotted in the second row of FIG. 1. Station 102b has computed a backoff interval 105b consisting of nine time slots. After four elapsed time slots, the backoff timer is frozen due to the transmission of said frame 101c by said station 102c. 
Each time the medium becomes idle, the station waits for a DIFS and then periodically decrements the backoff timer. The backoff timer of station 102b thus is only decremented again after the next DIFS period 104-2, where two time slots of the backoff interval 105b of station 102b elapse before the backoff timer is frozen again due to a frame 101d transmitted by station 102d. 
As indicated by the last row of FIG. 1, a data packet 103e arrives at station 102e during the transmission of frame 101c, so that station 102e has to start a backoff process as well. As can be seen by comparing the backoff intervals 105b of station 102b and 105e of station 102e after the next DIFS period 104-2, both backoff intervals 105b and 105e have the same length, so that, after the next DIFS period 104-3, three time slots elapse until both stations 102b and 102e concurrently start the transmission of frames 101b and 101e, respectively. If two or more stations start transmission simultaneously, a collision 106 occurs.
Unlike wired networks (e.g. with CSMA/CD in IEEE 802.3), in a wireless environment Collision Detection (CD) is not possible. Hence, a positive acknowledgment ACK 207 (see FIG. 2) is used to notify the sending station 202 that the transmitted frame 205 has been successfully received. The transmission of the ACK 207 is initiated at a time interval equal to the Short IFS (SIFS) 206-3 after the end of the reception of the previous frame 205.
If the acknowledgment is not received in a specified time interval, the station assumes that the transmitted frame was not successfully received, and hence schedules a retransmission and enters the backoff process again. However, to reduce the probability of collisions, after each unsuccessful transmission attempt the Contention Window is doubled until a predefined maximum (CWmax) is reached. After a successful transmission, the Contention Window is reset to CWmin.
After each frame transmission, a station must execute a new backoff process. Therefore at least one backoff is in between two transmissions of the same station.
In view of the above-mentioned problems, it is thus a further object of the present invention to improve wireless communications systems by reducing the number of collisions.
In radio systems based on medium sensing, a phenomenon known as the hidden-station problem may occur. This problem arises when a station is able to successfully receive frames from two different stations but the two stations cannot receive signals from each other. In this case a station may sense the medium as being idle even if the other one is transmitting. This results-in a collision at the receiving station.
To deal with the hidden-station problem, the IEEE 802.11 MAC protocol includes a mechanism based on the exchange of two short control frames, as depicted in FIG. 2: a Request-to-Send (RTS) frame 201 that is sent by a potential transmitter 202 to the receiver 203 and a Clear-to-Send (CTS) frame 204 that is sent by the receiver 203 in response to the received RTS frame 201. Said CTS frame 204 can be sent by the receiver 203 after waiting for a SIFS 206-1. If the CTS frame 204 is not received within a predefined time interval, the RTS frame 201 is retransmitted by executing the backoff algorithm described above. After a successful exchange of the RTS and CTS frames, the data frame 205 can be sent by the transmitter 202 after waiting for a SIFS 206-2. The implementation of the RTS packet 201 is optional, whereas all stations must be able to answer to a RTS frame 201 with the belonging CTS frame 204.
The RTS 201 and CTS 204 frames include a duration field that specifies the time interval necessary to completely transmit the data frame and the related acknowledgment (ACK) 207. This information is used by stations 208, 209 that can hear either the transmitter 202 or the receiver 203 to update their Network Allocation Vector (NAV) 210, 211, a timer that, unlike the backoff timer, is continuously decremented irrespective of the status of the medium. Since stations 208, 209 that can hear either the transmitter 202 or the receiver 203 refrain from transmitting until their NAV 210, 211 has expired, the probability of a collision occurring because of a hidden station 208, 209 is reduced. Of course, the drawback of using the RTS/CTS mechanism is an increased overhead, which may be significant for short data frames.
Furthermore, the RTS/CTS mechanism can be regarded as a way to improve the MAC protocol performance. In fact, when the mechanism is enabled, collisions can obviously occur only during the transmission of the RTS frame 201. Since the RTS frame 201 is usually much shorter than the data frame 205, the waste of bandwidth and time due to the collision is reduced.
However, when using the RTS/CTS mechanism, not only the hidden stations, but all stations in the coverage area of the transmitter 202 and the receiver 203 receiving said RTS frame 201 or CTS frame 204 update their NAVs 210, 211 and refrain from initiating further data transfers. This may result in a waste of bandwidth in particular if stations with spatially selective antennas are deployed in the WLAN system, because the spatially selective transmission and reception of frames naturally requires much less stations to be calmed down when trying to mitigate the hidden station problem.
Prior art document U.S. Pat. No. 6,611,231 B2 discloses methods, apparatuses and systems for use in a wireless routing network. One apparatus, for example, includes an adaptive antenna that is configurable to receive a transmission signal from a transmitter and in response, transmit corresponding multi-beam electromagnetic signals exhibiting a plurality of selectively placed transmission peaks and transmission nulls within a far field region of a coverage area. U.S. Pat. No. 6,611,231 B2 discloses to determine if there is a potential for interference with neighboring nodes prior to transmitting a CTS message, to transmit said CTS message to a targeted node using a narrow beam, if there is no significant potential for interfering with said neighboring nodes, and to transmit said CTS message to said targeted node and one or more of said neighboring modes using one or more beams if there is significant potential for interfering with said neighboring nodes.
This approach requires knowledge on the spatial propagation channels towards said neighboring nodes prior to transmission. Furthermore, said neighboring stations thus are intentionally calmed down by said CTS message to reduce interference, which may result in a waste of bandwidth. U.S. Pat. No. 6,611,231 B2 further discloses to have a pair or more of spatially separated wireless routing devices on a location or node. For example, a separation of about 20 wavelengths may be provided between the antenna arrays. The routing devices can allow a higher percentage of receive time using one of the antenna arrays, and also provide the potential of simultaneous transmit streams from the same approximate site.
In view of the above-mentioned problems, it is thus a further object of the present invention to improve the performance of wireless communication systems by enhancing the use of the available transmission bandwidth.